Bio
Swyx's defining contribution is the concept of the "AI Engineer" as a distinct professional identity, separate from both ML engineers (who build models) and traditional software engineers (who build deterministic applications). His Software 3.0 thesis argues that we are transitioning from Software 1.0 (code written by humans), through Software 2.0 (models trained on data, per Karpathy), to Software 3.0 where LLMs serve as the runtime and natural language becomes the programming interface. The AI Engineer sits at this intersection: a software engineer who builds AI-powered products using foundation models as building blocks, rather than training models from scratch.
Operating themes
- Operating thesis: We are in the transition from Software 2.0 to Software 3.0, where AI engineers -- not traditional software engineers -- will build the majority of new applications, and the key discipline is agent engineering: designing systems where LLMs control the application's logic flow.
- Agentic Ai Design
- Ai Business Models
- Ai Product Strategy
- Building With Llms
Cards
- We are in the transition from Software 2.0 to Software 3.0 — AI Engineers will build the majority of new applications — We are in the transition from Software 2.0 to Software 3.0 — AI Engineers will build the majority of new applications [Tier B]
Sources captured
- 2026-04 —
agent-engineering-latentspace.md(operator essay archive) - 2026-04 —
software-30-and-the-ai-engineer-landscape-talk-notes-slides.md(operator essay archive) - 2026-04 —
cognition-the-devin-is-in-the-details.md(operator essay archive)